Technology teach in. November 19, 2015 London

Transcription

1 Technology teach in November 19, 2015 London 1 FORWARD LOOKING STATEMENTS This presentation contains forward looking statements within the meaning of Section 27A of the US Securities Act of 1933, as amended, and Section 21E of the US Securities Exchange Act of 1934, as amended. These statements are subject to a number of risks and uncertainties that could cause actual results or outcomes to differ materially from those currently being anticipated. The terms outlook, estimate, project, plan, intend, expect, should be, will be, believe, trends and similar expressions identify forward looking statements. Factors which may cause future outcomes to differ from those foreseen in forward looking statements include, but are not limited to, competitive factors in the industries in which Reed Elsevier operates; demand for Reed Elsevier s products and services; exchange rate fluctuations; general economic and business conditions; legislative, fiscal, tax and regulatory developments and political risks; the availability of third party content and data; breaches of our data security systems and interruptions in our information technology systems; changes in law and legal interpretations affecting Reed Elsevier s intellectual property rights and other risks referenced from time to time in the filings of Reed Elsevier with the US Securities and Exchange Commission. 2 1

2 Strategic role of technology RELX Group Kumsal Bayazit Chief Strategy Officer, RELX Group Chair, Chief Technology Officer Forum 3 Agenda 4 Strategic role of technology RELX Group Kumsal Bayazit, Chief Strategy Officer Application of technology RELX Group Economics of an electronic decision tools business Government segment illustration Insurance segment illustration Q&A Vijay Raghavan, Chief Technology Officer, Risk & Business Information Mark Kelsey, Chief Executive Officer, Risk & Business Information Haywood Talcove, Chief Executive Officer, LexisNexis Government Solutions Bill Madison, Chief Executive Officer, Insurance Data Solutions Kumsal Bayazit, Chief Strategy Officer 4 2

3 RELX Group strategic direction Helping customers make better decisions, get better results and be more productive Where we are going Deliver improved outcomes to professional customers Combine content & data with analytics & technology in global platforms Build leading positions in long term global growth markets Leverage institutional skills, assets and resources across RELX Group How we are getting there Organic development: Investment in transforming core business; build out of new products Portfolio reshaping: Selective acquisitions; selective divestments Implications for business profile: Improving quality of earnings More predictable revenues Higher growth profile Improving returns 5 Revenue by format Reformatting print reference into electronic reference: largely done Transforming electronic reference into electronic decision tools: current focus 14% 22% 21% 19% 18% 27% 25% 37% 33% 17% 51% 52% 64% 58% 56% 64% 60% 15% 16% 15% 15% 14% 14% 17% 15% 12% 13% 12% 12% 69% 12% 64% 66% 66% 61% 63% 59% 14% 14% 48% 50% 37% 30% 32% 35% 28% 22% 22% Print: Print to electronic migration largely complete Face to face: continuing to grow Electronic: Transitioning from electronic reference to electronic decision tools adding broader data sets more sophisticated analytics leveraging more powerful technology H1 Print Face to face Electronic 6 3

4 Electronic reference to electronic decision tools Improved outcomes, demonstrable and measurable customer value Electronic reference Electronic decision tools Reference: search and retrieve critical information Decision support tools: analysis of content, datasets, facts and patterns to help make a decision Embedded machine to machine decision tools Built in tools Stand alone tools 7 Scientific, Technical & Medical (STM) Illustrations of decision support tools Built in tools on ScienceDirect 3D Viewer SciVal Stand alone tools Virtual Microscope Pathway Studio Interactive Plots Reaxys 8 4

5 Legal Illustrations of decision support tools Built in tools on LexisAdvance Legislative Outlook Stand alone tools MedMal Navigator Research Map Verdict and Settlement Analyzer Legal Issue Trail Market Tracker 9 MedMal Navigator legal medical litigator Illustration of decision support tools 10 5

6 MedMal Navigator legal medical litigator Standard of care 11 MedMal Navigator legal medical litigator Case value assessment 12 6

7 MedMal Navigator legal medical litigator Expert witness 13 Exhibitions Illustration of early pilots Integrating diverse data with different update cycles Creating a rich customer profile Exhibitor sales and event history (daily) Online research/planning portal usage (daily) Attendee registration and demos (daily) Mobile app usage (postevent) Event surveys (post event) Sales leads from event (post event) Exhibitor marketing services data (post event) Visited event for 2 days; attended multiple sessions Is somewhat satisfied with event; has high need John Q. Customer Attends every other year Actively researches exhibiting companies / segments prior to event Logged in 5 times; searched specific companies Drive better matchmaking Predict attrition 14 7

8 Our four key capabilities to deliver electronic decision tools 1. Deep customer understanding 2. Leading content and data sets 3. Sophisticated analytics 4. Powerful technology in global platforms Deep customer understanding Scientists and doctors Risk professionals Lawyers Exhibitors c10m researchers 1.8m authors 17,000 editors 700,000 reviewers >10m monthly unique visitors 100% of US P&C insurers Top 50 banks Transactions per annum: >300m user >3bn machine to machine >400m batch Top 250 Global Law firms AmLaw 200 c75% of Fortune 500 c1m subscribed users >250m searches per year c2bn document views >500 events >7m event participants >140k exhibitors >40 sectors 16 8

9 2. Leading content and data sets Scientific, Technical & Medical >1m article submissions per year >13m articles on Science Direct 16% of global research c2,500 journals Global chemical compound and reaction databases (>50m compounds; >35m reactions) US drugs database (>100k drug packages) Global disease pathways (>5m molecular facts) >1bn miles of UK driving behaviour >100m US bankruptcy records >450m US criminal records US medical providers (c6.5m entries) >4bn US motor vehicle registrations Global watchlists (2m+ entries; >200 jurisdictions) >9bn unique US personal identifiers Risk & Business Information Global air fleet specifications (>350k aircraft; >100k daily flights) >1.5bn documents 4bn legal topic links Primary law from over 150 countries >500 events in 30+ countries Legal Global business news (c30k sources; 57 languages) >20m briefs, pleadings, motions, jury instructions etc. >1m jury verdicts and settlements Global patents (>100 patent offices, >100m records) >7m event participants c3.5m m2 net space sold in 2014 >140k exhibitors Exhibitions Sophisticated analytics 18 Entity resolution, linking and clustering Scoring models and attributes Descriptive, diagnostic, predictive and prescriptive analytics Visualisation to represent clusters, links and insights 18 9

10 4. Our technology capabilities Several technology hubs across North America, Europe and Asia c50% software engineers $1.3bn technology spend c20% of operating cost base $500m average annual capex >95% spent on technology 19 Summary Combining content & data with analytics & technology is at the core of our strategy Decision tools add significant value to our customers and to our business We are migrating reference to decision tools across the whole of the RELX Group 20 10

11 Application of technology RELX Group Vijay Raghavan Chief Technology Officer, Risk & Business Information 21 RELX Group approach to technology Technology agnostic open sourced, third party and proprietary Algorithms are the secret sauce Re use approaches and technologies across the Group Attract and retain best talent 22 11

12 What do we mean by technology at RELX Group? Primary research Public records Entity resolution Proprietary data News articles Link analysis Big Data Contributory databases Unstructured records Structured records Refinery Linking Fusion Clustering analysis Complex analysis Unstructured and structured content Big Data platforms Analysis applications Sample capabilities Over 3 petabytes of content 10s of billions of records 100s of thousands of sources Billions of unique name & address combinations Grid computing with low cost servers Data centric languages (brings to code to the data) Linking algorithms that generate high precision and recall Machine learning algorithms to cluster, link and learn from the data Integrated delivery system for high speed data fabrication without compromising high speed data retrieval Scoring models & attributes Social graphs to identify patterns Visualisations to represent clusters, links & graphs of entities Scientific author disambiguation Recommendation engines Identity verification Fraud detection and prevention Case outcome prediction Know your customer 23 Example: LexisNexis Risk Solutions (LNRS) What questions does LNRS answer for its customers? Are you who you say you are? 2. Who else might you be, or claim to have been in the past? 3. What other people and/or assets are associated with you? 4. What kind of a risk are you in a given context? a) in calculating your insurance premium, or b) in processing your claim, or c) in granting you access to credit, or... d) in doing business with you as a vendor or customer 5. Can I quantify the risk that you represent in the form of a score? 6. Which of these millions of transactions should I look at in case there s something suspicious? 7. What small subset of these thousands (or millions) of events have something in common which will cause me to look more closely without wasting time on false positives? 24 12

13 It all starts with the data! 25 Break down of record counts for the more popular data sets: Data source # of records Data source # of records Associates/relatives 1.8 billion Private phones 172 million Bankruptcy 111 million Professional licenses 94 million Business BDID's 283 million Property 2.5 billion Business people links 959 million Sex offenders 550,000 Canadian phones 62 million SSN's 7.2 billion Consumer header 10.8 billion Student records 38 million Criminal 450 million TIN 2.9 million Date of birth 5.2 billion Unique ADLs active 257 million Death 98 million Utility 645 million Drivers licenses 397 million Vehicle titles 635 million EDA phones 124 million Vehicle registrations 4 billion FEINs 10.4 million White pages 116 million Historical phones 800 million Wireless phones 101 million Hunting and fishing licenses 67 million Yellow pages 14 million Liens and judgments 244 million etc. People at work 1.5 billion 25 So why is this a Big Data problem? 26 Candidate File 20 billion records Match by date of birth and locale 900 million matches 20 hours of joins across 400 machines Every possible match is scored for specificity Amounts to about comparisons Match Candidates 26 13

14 After we link records, we can then create unique insights Source Data Matched Data Sets Scores & Attributes Tax liens, felonies, bankruptcies Professional licenses Landline & cell phone Bureau header data Property deeds Court judgments Voter registration Other (education, etc) Compiled records for more than 250m identities Primary attributes Summarise a particular characteristic of a consumer Examples include: Number of addresses Tax assessed residence value Composite attributes Summarise aggregate consumer behaviour (wealth, income, mobility) Created by combining primary attributes Scores Algorithms are our secret sauce... Industry specific scores Custom scores 27 What do we mean by algorithms are our secret sauce? An example (SALT) 28 A single source of data is insufficient to overcome inaccuracies in the data A single data source view the holes represent inaccuracies Instead, absorb data from multiple data sources and link it very accurately using a probabilistic linking algorithm A multiple data source view the holes in the core data have been eliminated using field value specificities to derive the probability of a match 28 14

15 It might sound easy, but it is an evolution over time 29 1 Data centric approach 2 Data flow oriented Big Data technology 3 Data disambiguation & linking technology 4 Data graph processing technology 5 Real time enabled technology 6 Visualisation & workflow technology Our technology platform is Better, Faster, Cheaper 30 Analytics and algorithms War chest of algorithms, written in a data scientistfriendly language (ECL) Better: Advanced linking ability improves data quality and precision; reduces false positives. Our algorithms are our secret sauce. Produces superior products Big Data technology (HPCC) Open sourced Big Data platform that underpins a substantial portion of RELX Group revenue Faster: High speed of processing and response; rapid development environment that frees up resources Solves more problems in less time Hardware and operating system Parallel computing on commodity servers and Linux OS, augmented by other open source modules as needed Cheaper: Combination of commodity hardware, high level language, and extremely effective algorithms requires fewer people and resources Reduces technology expense 30 15

16 We use our Big Data platform across RELX Group 31 ScienceDirect Advanced Recommender doubled the click through rate SciVal platform RealPulse Lexis Advance Media Neutral Content Repository data process time cut from days to hours Content enrichments Relationship engine Empower + Engage uncovered attrition factors, engaged at risk exhibitors Engagement with external technology community drives further innovation RELX Group Community Partners 32 16

17 Our strong technology talent drives innovation 33 Product innovation driven by the businesses Innovation at the tooling level (for developers/modelers) Innovation at the algorithms/languages level Innovation at the platform level Innovation at the hardware level Technology innovation to drive time to market The attrition within LN Risk Solutions is less than 7% Employee satisfaction scores are in the 70% range The average tenure of a technologist at Risk Solutions is 10 years 33 Summary Expert in using technology to solve Big Data problems Embedded in Risk & Business Information, extending across RELX Group Deep talent pool leveraged across RELX Group 34 17

18 Economics of an electronic decision tools business Mark Kelsey Chief Executive Officer, Risk & Business Information 35 Risk & Business Information: a key part of RELX Group 36 Revenue H underlying growth: +3% +6% Adjusted operating profit H underlying growth: +5% +3% +1% Legal Exhibitions Scientific, Technical & Medical +2% +8% Legal Exhibitions Scientific, Technical & Medical +5% Risk & Business Information Risk & Business Information +7% +7% 36 18

19 Leader in data and analytics that enable customers to evaluate risk and support key decisions $2.4bn* revenues with strong growth drivers Margins of 35%* Low capital intensity capexc4% of sales Market leading data, technology, and proprietary analytics by format Face to face 2% Print 7% Electronic 91% Risk & Business Information revenue proforma 2015 continuing businesses by geography Rest of world 5% Europe 16% North America 79% Subscription 34% by type Advertising 3% Transactional 63% *FY Risk & Business Information revenue by segment Proforma 2015 continuing businesses Government & Health Care Other magazines and services Major Data Services Business & Data Services Insurance Solutions Our customers c100% of US P&C insurance carriers All top 100 global banks c90% of Fortune 500 All 50 US states c70% of US local governments c80% of US federal agencies 38 19

20 Risk & Business Information underlying revenue growth 39 7% 6% 6% 5% 3% H Risk & Business Information underlying revenue growth contribution 40 7% 5% 6% 6% Base market growth contribution 3% Contribution from recent product introductions* H % 32% 35% 35% 35% Adjusted operating margin * Less than 5 years old 40 20

21 Strategic priorities for driving organic growth Core markets Continuous product innovation to improve customer outcome; effectiveness, efficiency and compliance Drive deeper into innovative applications and increase penetration across customer workflows Adjacent markets Pursue growth in attractive adjacent markets where our core strengths can be leveraged International Address international opportunities in selective geographies: leveraging skill sets, technology, analytics and experiences 41 This slide is intentionally left blank 42 21

22 Government segment illustration Haywood Talcove Chief Executive Officer, LexisNexis Government Solutions 43 LNRS provides analytics solutions to solve problems across US government segments Our focus includes: Stopping fraud and catching criminals Making customers more efficient in a challenged budgetary environment Solutions cost around 1% of savings We use: Deep customer insight Unique data content Visual analytics Leading technology 44 22

23 Who we serve out of 50 States Dept. of Revenue 48 out of 50 States Dept. of Health & Human Services All executive level federal agencies Top 100 largest state & local US law enforcement agencies Public Safety Intelligence Agencies Tax & Revenue Health & Social Locate person of interest Identify asset ownership Find non obvious relationships Detect patterns and hidden relationships between persons of interest Conduct cyber forensics investigations Monitor insider threat Prevent tax fraud and identify tax evasion Discover new sources of revenue by detecting tax fraudsters and nonfilers Prioritise collection efforts Ensure welfare program integrity Validate claimant identity Assess claimant eligibility Monitor payments to the right individual LNRS Government solves complex government identity and fraud problems New York Police Department Using Accurint for Law Enforcement plus for solving Cold Cases, it makes difficult and time consuming tasks easy and fast, and it s a fantastic lead generator. Indiana Department of Revenue Using the TRIS* program without slowing down the refund process, allows us to identify those culprits, Alley said. It's increased our efficiency and enabled us to pursue other forms of tax fraud. The use of the LexisNexis technology in 2014 has stopped $42 million in refunds being mailed to criminals. New Jersey Department of Labor & Workforce Development Approximately 646 instances of attempted identity fraud have been prevented using the LexisNexis Identity Management solutions. LexisNexis has helped New Jersey Department of Labor stop $4.4m in unemployment insurance fraud over the past 17 months. *Tax refund intercept system 46 23

24 Customer use case: Mississippi National Accuracy Clearinghouse (MS NAC): Solving a key benefits fraud problem 47 Online government applications for food stamps are at an all time high More than 48m Americans have been on the program for 35 straight months Now one in every five Americans are enrolled on the program Pre paid / electronic benefit transfer (EBT) cards Benefit dollars paid to Pre paid/ebt Cards has doubled each year to improve turnaround Every American s identity has already been stolen Over 850m names, SSNs and bank accounts stolen in record breaches since 2005 MS NAC Set of five US States expressed need to validate that individuals applying for food subsidy benefits in one state were not also applying for benefits in another 47 Customer use case: MS NAC solution leverages innovative LNRS technology and analytics 48 Our program combines customer data from states with our public records data to identify fraudulent patterns, including applications for benefits in multiple states 48 24

25 Customer use case: MS NAC: contributory data handling and build processing 49 >96m records ingested per annum c24m queries processed per annum >70% reduction in duplicated participants; savings of c$150 million. due to increased identification of fraud 49 Our longer term vision is to build a Contributory Solution for every Health & Human Services (HHS) segment to solve bigger fraud problems A SuperNAC Medicaid NAC TANF 1 NAC DMV 2 NAC Unemployment Insurance NAC Food Stamps NAC Public Housing NAC Visual Analytics XML Portal Batch Our VISION HPCC Technology Contributory Data 1. TANF: Temporary Assistance for Needy Families 2. DMV: Department of Motor Vehicles 50 25

26 Insurance segment illustration Bill Madison Chief Executive Officer, Insurance Data Solutions 51 LNRS provides vital contributory & external data and analytics to the US insurance industry Our expertise and product solutions enable carriers to: Make better and faster risk underwriting decisions Make more accurate policy pricing decisions at issue and renewal Reduce claim losses using advanced data & fraud detection analytics Streamline the customer application and policy management process 52 26

27 Insurance Solutions growth has been driven by innovation 53 c$900m Insurance revenue c$30m CLUE/Auto MVR CLUE/Property A.D.D. CLUE/Auto MVR CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR InsurQuote FIRSt Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Auto Data Prefill InsurView ODG MVR Predictor Model Commercial Data Solutions InsurQuote FIRSt Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Marketing (2009) Marketing Prefill Property Data Prefill ATTRACT 2.0 Foreclosure Life Non Credit DHDB 2.0 Auto Data Prefill InsurView ODG MVR Predictor Model Commercial Data Solutions InsurQuote FIRSt Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Claims Datafill UK Public Records Driving Beh. Models Life Prefill Commercial CLUE Commercial Prefill CLUE Enhanced MVR Retention Models Marketing (2009) Marketing Prefill Property Data Prefill ATTRACT 2.0 Foreclosure Life Non Credit DHDB 2.0 Auto Data Prefill InsurView ODG MVR Predictor Model Commercial Data Solutions InsurQuote FIRSt Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Telematics Claims Compass Claims Medical Hist. Citizen Reporting Workers Comp Policy Data UK No Claims Database Wind/Hail Peril Models Life Persistency Models Claims Datafill UK Public Records Driving Beh. Models Life Prefill Commercial CLUE Commercial Prefill CLUE Enhanced MVR Retention Models Marketing (2009) Marketing Prefill Property Data Prefill ATTRACT 2.0 Foreclosure Life Non Credit DHDB 2.0 Auto Data Prefill InsurView ODG MVR Predictor Model Commercial Data Solutions InsurQuote FIRSt Current Carrier Activity Files Policy Watch ATTRACT Credit CP Rules Point of Sale (POS) Driver History Database CLUE/Property A.D.D. CLUE/Auto MVR Deeply embedded at the point of underwriting, we are focused on driving deeper integration across all parts of the carrier workflow Workflow: MARKETING CONTACT QUOTE QUOTE UNDERWRITING RENEWAL COMPLIANCE CLAIM AUTO: System to system integration level: Opportunity Growing High High Growing Growing Growing Growth Growth HOME/LIFE/COMMERCIAL: System to system integration level: Opportunity Growing Growing Growing Growing Growing Opportunity Growth Growth 54 27

28 Data assets available across the insurance continuum MARKETING CONTACT QUOTE QUOTE UNDERWRITING RENEWAL COMPLIANCE CLAIM Carrier / agent system Value added analytics Overlay additional info to further refine results e.g., Insurance Score Single point of entry standard XML schema Pull on insurance data infrastructure Output tailored to insurer s systems, pricing and risk criteria Insurance data infrastructure Contributory data: Policy and loss history data Police reports: ecrash, CRU and Coplogic Public records data: Driver history, vehicle data, bankruptcies, liens, judgments, court records, census data, etc. State gateways: Motor vehicle Credit bureaux: reports Credit bureaux & business information 55 Machine to machine integration: the power of analytics 56 MARKETING CONTACT QUOTE UNDERWRITING RENEWAL COMPLIANCE CLAIM Today LexisNexis manages over 400 models within the infrastructure Industry models and custom models Models creates a platform for today s needs and tomorrow s advancement Scores and reason codes Data attributes The platform advances and evolves to the needs of the market Signal data sourced models Multi variant decision processes 56 28

29 The power of data: process and risk assessment 57 MARKETING CONTACT QUOTE UNDERWRITING RENEWAL COMPLIANCE CLAIM There will always be interest in solutions that provide better risk assessment. Motor vehicles records Claims history Profiles of the household Financial history (credit) Policy data Marketing segmentation Driving behaviour Telematics We are constantly adding more data 57 Telematics made real through analytics:experts predict the first self driving cars may be on the roads in 2016 Today we manage: Over 1 billion vehicle miles 120 earned vehicle years Why is this important? This information provides a baseline of knowledge It provides market learnings, not just data. By building a database of all posted speed limit information, you can now understand simple things like on ramp / off ramp driving behaviours Analytics provides a way to transfer data and market knowledge to actionable intelligence 58 29

30 59 The extended principle of data Insurance Solutions 59 Using data to enhance the insurance experience 60 MARKETING CONTACT QUOTE UNDERWRITING RENEWAL COMPLIANCE CLAIM Industry challenges The industry is spending over $6bn a year to excite consumers The process was more about data collection Customer experience was poor at best not to mention data errors when entering data into the carrier system, requiring them to revert to the customer LNRS solutions Auto Data Prefill: Provides critical household information (drivers, vehicles and existing coverage) needed to quote and underwrite a policy, with minimal input required (name, address and date of birth) Market Success Stories with Prefill Reduce quote time from over 20 minutes to less than 5 minutes Improve information product order hit rate by as much as 8% Reduction of bad VIN s from 12% to 2% over a 12 month period of time Call Centre first time close rate increased by 60% in the first 30 days of launching the service 60 30

31 Prefill: The next generation = Data Prefill for Mobile Mobile Auto Data Prefill Welcome to Data Prefill for Mobile Get a personalized insurance quote today AUTO HOME BOTH By simply taking a picture of the barcode on the back of a Driver License, LexisNexis can prefill the application components; with no data entry

32 63 David Jones Stephanie Jones 64 32

33

34 Strategic priorities for driving growth 67 Core markets Continuous new product innovation to improve economic outcomes of insurance carriers and intermediaries Increase penetration across carrier workflows, from marketing and point of contact through underwriting to claims Adjacent markets Pursue growth in attractive adjacent markets leveraging LNRS core strengths (e.g. Life, Home, Commercial) International Address international opportunities in selective markets leveraging skills sets, technology, analytics and experience 67 Summary Combining content & data with analytics & technology is at the core of our strategy Decision tools add significant value to our customers and to our business We are migrating reference to decision tools across the whole of the RELX Group 68 34

35 Technology teach in November 19, 2015 London 69 35